IRJET- Investigation into use of Soft Robotic End Effector for Assistive Gripping

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 05 Issue: 03 | Mar-2018 www.irjet.net p-ISSN: 2395-0072
Investigation Into Use of Soft Robotic End Effector For Assistive
Gripping
Girish Raghunathan1, Venkatesh Raghunathan2, L Krishna Reddy3, Basil Rajan4
1,2,3,4 Student, Department of Mechanical Engineering, Christ Deemed to be University, Bangalore, Karnataka, India
----------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Researchers and scientists alike have been
searching for inexpensive, robust, and a comparable
substitute which integrate the functionality, compliance,
versatility, and strength of human arms. Some measure of
progress has been achieved in case of metal or alloy made
prostheses or orthotics, composite layered bionic arms etc.
However these models are hard to design and costly to
make. Also, though they exhibit excellent strength
characteristics their range of movements is to a major
degree constrained and limited. In lieu of such difficulties,
the largely untapped area of soft robotic elements offers a
feasible and viable solution. Materials such as soft elastomer
and rubbers on integration with other available materials
such as paper, cardboard or light plastic can be easily
fabricated into a variety of designs to suit the requirements.
The concept of PneuNet - an internal air channel or
pneumatic network which can deform in a desired direction
when restrained in a certain way is justifiably applicable in
such situations. Rubber incorporated with PneuNets share
similar compliance of human arms and demonstrates the
same natural manner of deformation. Moreover, these are
relatively inexpensive to prototype and are extremely
versatile. In our present work, we have attempted to design
and fabricate a soft robotic end effector which can primarily
act as assistive prostheses. We also introduce an electronic
control system based on flex sensors and the Micro-
controller Arduino Mega 2560 coupled with the simple
pneumatic circuit. The air flow control is achieved by a
simple modified servo valve design which will also be
discussed.
Key Words: Compliance, PneuNet, Assistive
Prostheses, Micro-controller, Flex Sensor, Soft Robotic,
End Effector
1. INTRODUCTION
The area of soft robotics being applied in the field of
medicine as substitutes for various organs of the body has
been extensively researched. Most of the actuators that
have been synthesized till now make use of the expansion
and contraction property of elastomers mainly for cardiac
assist devices. Actuators have also been made to replicate
the muscular motions to some accurate degree. PneuNet
actuators have been primarily made to simulate the finger
motion giving some compliance to its motion.
Prototype models have been made with multiple PneuNet
fingers to clone hand motion. For the real time dynamic
control implantable electrodes have preferred, though
piezo sensors, flex sensors, shape memory alloys etc. have
also been looked at. Usually these sensors are integrated
with a microcontroller like Arduino Uno or similar and
input values being programmed to it. A range of materials
including silicone elastomers, polyurethanes, plastic resins
and shape memory alloys have been successfully
incorporated into working prototypes and tested.
However these soft structures have yet to improve upon
their strength characteristics though they do possess a
higher degree of compliance.
2 Concept of PneuNet
A PneuNet is an abbreviation for pneumatic network
though more precisely it denotes an internal enclosed
channel for its activation by the passage of compressed air
ranging from 90kPa to more than 300kPa. These so called
PneuNets gained attention with its first conceptualization
by the Whitesides group working at the Bio lab at Harvard.
From its inception, it has always been seen as a potential
substitute for human tendons and muscles attributed
mainly to its comparable compliance to that of a generic
human muscle.
The construction of a PneuNet is an extremely simple
process; all that is needed is to sandwich two thin flexible
elastomer planes on top of one another and seal the edges
off so that they are air tight. If pressurized air is allowed to
flow into the internal orifice it pushes the soft rubbery
layers from the inside so that they bulge out in accordance
with the direction or mode in which the layers are
constrained. The expanding layer deforms as a
consequence of the work done by the pressurized air on it.
This deformation may be constructively utilized in two
ways, namely:
1. By constricting the deformation such that the final shape
shifting of the elastomer occurs anisotropically.
2. By connecting elements to the deforming rubber layer in
a manner that they are actuated by it.
PneuNets may be created by utilizing various available
techniques some of them being soft lithography, stereo
lithography, elastomer casting, flexible 3-D printing etc.
PneuNets can be synthesized in various configurations,
shapes modes and designs to simulate, bend, actuate or
deform exactly according to the required specifications of
displacement, angle, velocity and force by merely
constraining the corporeal structure and varying the
supply pressure.
© 2018, IRJET | Impact Factor value: 6.171 | ISO 9001:2008 Certified Journal | Page 970

This makes PneuNets a fascinatingly versatile piece of
hardware with potentially limitless applications across
myriad field including those of medicine, soft robotics, food
automation, inexpensive robots, orthotics, non-destructive
inspections, pick and place operations and many more.
3 Sizing and Dimensioning of the Gripper
The average size of the human hand is naturally
designed to achieve the highest functionality. The relative
volumes occupied by the palm to the sizing of the fingers
and the extra indented orientation of the thumb also
provides for the best combination of factors that is
required for all hand related operations. These
considerations were taken into account and hence the
following choices were made. The weight of the whole
actuator does not exceed 438.6 g and the thumb is
proportionally smaller and is suitably oriented. the
PneuNet themselves are 24 mm x 18 mm with a 12.5 mm
top curvature radius and the internal air channel is about
1mm in thickness. The whole length of one actuator
(finger) is about 178.45mm long. This slightly exaggerated
length was chosen to study the PneuNet operation with
more precision. The molds for the finger actuators were
adapted from standard available designs from the World
Wide Web.
Fig -1: Soft Robotic End Effector
4. Modeling
The model was prepared using Solidworks 2016 in a 3
dimensional form for ease of simulation and analysis. The
various internal air channels are represented and the
model is slightly simplified to ease up the simulation
process. Care is taken to accurately represent each and
every feature so that the structure of mold, placement of
cores, sizing, and overall multi-part assemblies may be
clearly visualized. After the 3-D model is created, each file
representing a mold which is to be 3-D printed is converted
into a .stl file. The various parts of the soft robotic end
effector which were CAD modeled are subdivided into two
main categories namely:
1) Parts which form the mold
2) Parts which are form the finger fixtures
3. Simulation and FEM Analysis of soft robotic
finger
To gauge the behavior of a PneuNet, a rudimentary FEA
was carried out on Solidworks 2016. For the simulation 3
conditions were chosen which are:
• A single PneuNet was sketched and the various stresses
and strains acting on it for a variety of loads were tested.
• A simplified model of finger with PneuNets incorporated
was modeled and analyzed for whole body behavior
• Another conceptual simplified model was tested with
Simulia-Abaqus, a standard FEM Package for accurate
deformation rate and stresses acting on it.
• The general steps that were followed for the analysis of a
single PneuNet in Solidworks Simulation were:
1. CAD Modeling of the structure of PneuNet or PneuNet
finger in the software package
2. Assigning material properties to the structure. In our
case, rubber was the material used. The properties of
rubber chosen were based on its average exhibited
properties.
3. The body is then assigned fixture conditions so that the
deformation about the fixture may be determined.
4. External and Internal Loading: The various loads acting
on the body are assigned next. For our analysis, a constant
internal pressure loading condition of 1 psi was selected
and applied.
5. Mesh: Mesh size was selected as fine mesh for precision
and clarity in final results. The mesh parameters were
rounded of for uniform mesh size. Triangular mesh was
selected for simplicity of analyses.
6. The study was then run and the results were then
tabulated and conclusions drawn.
Fig -2: SolidWorks simulation of the PneuNet based finger

5. Fabrication Details
• The molds for the soft robotic finger components were
generated by 3-D printing as elucidated before.
• The elastomer solution was prepared by thoroughly
mixing the part A and part B constituents of the rubber
solution together in the prescribed 1:1 ratio.
• The stirred solution was then poured into the molds
slowly and with caution keeping in attention to uniformly
fill all the cavities with rubber solution.
• The casting process was repeated for all the 4 other
finger-molds, 2 thumb molds, 1 thumb pad, 1 palm skin and
1 palm pad.
• The curing time for Dragon Skin 20(Smooth-On Inc.)
ranges from 4-6 hours during which the solution must be
kept in a relatively cool and undisturbed place.
• After the curing was completed the molds were broken
apart in order to obtain the final soft robotic part.
• One advantage of the used rubber was that the use of a
release agent to facilitate product removal was efficiently
eliminated.
• The assembly of the soft robotic gripper followed the
completion of the molding process.
• Each soft robotic finger component was connected to its
corresponding base bone part using cyanoacrylate solution
for strong adhesive bonding.
• 1/16” tubes were inserted in the supply channels of each
part to facilitate compressed air supply.
• The finger bone sub assembly themselves were bolted to
the main palm and the same procedure carried out with the
thumb.
• The thumb finger base bone has a pivoted connection to
enable slightly free movement.
• The said pivoted joint was elevated in correct position
using some pieces of black anti-static foam.
• The palm pad and thumb pad were glued in place and the
skin inserted using strong adhesive bonding solution.
• The supply tubes themselves were connected to 3-way
and 4-way connectors and finally joint to a common air
supply manifold.
6. Results and Discussions
After the aforementioned simulation experiments were
conducted and the results tabulated.
The following points were noted as points of interest:
1. We note that as the load applied on the finger (weight of
the object being lifted) by the finger the bend angle shows
a decreasing trend. This stands to logic as more the load
applied equally greater will have to be the input pressure.
Chart -1: Graph of weight carried vs Bend Angle for Index
Finger
Chart 2: Graph of weight carried vs Bend Angle for Middle
Finger
Chart -3: Graph of weight carried vs Bend Angle for Ring
Finger
Chart -4: Graph of weight carried vs Bend Angle for Little
Finger

Chart -5: Graph of weight carried vs Bend Angle for
Thumb Finger
2. The tip force of each finger depends upon the amount of
input air discharge applied which increases in proportion.
It also is observed that the size of the finger and the
number of PneuNets per finger. Hence we see that the tip
force exerted by the thumb is relatively less than that of
the other fingers.
Table -1: Tip Force values of each finger
3. The theoretical von Mises stresses, strains, and
displacements for the solitary PneuNet, the finger actuator
and the simplified model is obtained from the software
simulation. The mode of operation and expected trajectory
of motion may also be viewed.
Table -2: Values of Parameters For a Single PneuNet
For input pressure of 1 psi obtained from
Solidworks Simulation 2016
Table -3: Values of Parameters For a PneuNet finger
For input pressure of 30 psi obtained from
Solid works Simulation 2016
4. From among the bend angle values obtained, it is found
out that the bend angles for the little finger were the
greatest of the five while the thumb was expectedly the
least for the same load. The case of the thumb finger is
easily explainable as it has the lowest number of
PneuNets. The variation of bend angles in case of the other
fingers is mainly due to the difference in the amount of
boundary separation of consecutive PneuNets. More the
boundary separation within the limit greater is the space
of expansion.
5. The attempt involving the incorporation of flex sensor
for electronic control was partially successful. The servos
operated are able to respond to the signals transmitted by
the microcontroller as the result of flexure of the flex
sensors but results in a lag of operation. This is attributed
to the low sensitivity of the flex sensor, and certain play in
the connections and its length. It is suggested that the
same experiment may be carried out using a longer flex
sensor.
Fig -3: Soft Robotic end effector with the flex sensor
control glove
6. The experiments were conducted with a pump with a
net discharge of 21 lpm. It was observed that for sustained
and satisfactory bending an average discharge of about 1.5
to 1.8 times the mentioned value. It is presumed that
higher bending may be observed with higher discharges of
air. Also the pressure required for bending a finger
actuator is about 30 psi which is also borne out by the
literature.
7. On the basis of tip force and bend angle values obtained
it is noticed that sufficient grip force and orientation to
successfully grasp any object. The lifting capacity may be
increased by increasing input air pressure and discharge.
7. CONCLUSIONS
Some of the conclusions made from this project are:
• PneuNets make for a viable and feasible idea for
inexpensive body organ actuator.
S NO. FINGER TIP FORCE (N)
1 Thumb 0.84366
2 Index 1.1962
3 Middle 1.03986
4 Ring 1.3734
5 Little 1.4715
Parameter Value
Maximum Von Mises Stress(N/mm2) 2.254e+005
Minimum Von Mises Stress(N/mm2) 2.283e+002
Maximum Displacement (mm) 2.341e+000
Minimum Displacement (mm) 1.000e-030
Parameter Value
Maximum Von Mises Stress(N/mm2) 1.636e+008
Minimum Von Mises Stress(N/mm2) 2.716e+002
Maximum Displacement (mm) 1.015e+005
Minimum Displacement (mm) 1.000e-030

• Actuation of soft actuators can be accomplished with low
air pressure ranging from 0.1 to 5 bars.
• Deformation of PneuNet actuator is sufficient to cover
the range of finger motion.
• Stresses developed in actuator during operation are well
within the materials’ yield stress values.
• The gripping is limited currently to light weight objects.
• With appropriate attachments and patient tailored hand
size the PneuNet actuator may be succefully deployed as
an assistive device for gripping objects.
8. SCOPE FOR FUTURE WORK
Some of the modifications which may be suggested for
further improvement of the project are as follows:
• The sizing of the actuator could be made more compact
by properly design the internal air channels.
• Other sensors or electrode based control elements may
be introduced for better control.
• The addition of stiffening elements to the soft finger to
reduce wobbling effects can be taken up.
• Lighter material substitutes for soft finger may be looked
into.
• Miniature air compressors can be designed to supply
power while reducing the weight.
• A complete attachment device to which the soft end
effector can be connected and the other end be strapped in
place of the organ.
REFERENCES
[1] K. C. Galloway, P. Polygerinos, C. J. Walsh, and R. J.
Wood, “Mechanically programmable bend radius for fiber-
reinforced soft actuators,” 2013 16th International
Conference on Advanced Robotics (ICAR), 2013.
[2] D. Holland, E. Park, P. Polygerinos, G. Bennett and C.
Walsh, ”The Soft Robotics Toolkit: Shared Resources for
Research and Design” Soft Robotics, vol. 1, no. 3, pp. 224-
230, 2014.
[3] Z.Wang, P. Polygerinos, J. Overvelde, K. Galloway, K.
Bertoldi and C.Walsh, ”Interaction Forces of Soft Fiber
Reinforced Bending Actuators”, IEEE/ASME Transactions
on Mechatronics, vol. 22, no. 2, pp. 717-727, 2017.
[4] B. S. Homberg, R. K. Katzschmann, M. R. Dogar, and D.
Rus, “Haptic identification of objects using a modular soft
robotic gripper,” 2015 IEEE/RSJ International Conference
on Intelligent Robots and Systems (IROS), 2015.
[5] C. Majidi, ”Soft Robotics: A Perspective—Current
Trends and Prospects for the Future”, Soft Robotics, vol. 1,
no. 1, pp. 5-11, 2014.
[6] A. Marchese, R. Katzschmann and D. Rus, ”A Recipe for
Soft Fluidic Elastomer Robots”, Soft Robotics, vol. 2, no. 1,
pp. 7-25, 2015.
[7] P. Polygerinos, K. C. Galloway, S. Sanan, M. Herman, and
C. J. Walsh, “EMG controlled soft robotic glove for
assistance during activities of daily living,” 2015 IEEE
International Conference on Rehabilitation Robotics
(ICORR), 2015.
[8] N. Bartlett, V. Lyau, W. Raiford, D. Holland, J. Gafford, T.
Ellis and C. Walsh, ”A Soft Robotic Orthosis for Wrist
Rehabilitation1”, Journal of Medical Devices, vol. 9, no. 3, p.
030918, 2015.
[9] C. T. Oneill, N. S. Phipps, L. Cappello, S. Paganoni, and C.
J. Walsh, “A soft wearable robot for the shoulder: Design,
characterization, and preliminary testing,” 2017
International Conference on Rehabilitation Robotics
(ICORR), 2017.
[10] R. Martinez, C. Fish, X. Chen and G. Whitesides,
”Elastomeric Origami: Programmable Paper-Elastomer
Composites as Pneumatic Actuators”,Advanced Functional
Materials, vol. 22, no. 7, pp. 1376-1384, 2012.
[11] R. Niiyama, X. Sun, C. Sung, B. An, D. Rus and S. Kim,
”Pouch Motors: Printable Soft Actuators Integrated with
Computational Design”, Soft Robotics, vol. 2, no. 2, pp. 59-
70, 2015.
[12] B. Peele, T.Wallin, H. Zhao and R. Shepherd, ”3D
printing antagonistic systems of artificial muscle using
projection stereolithography”, Bioinspiration &
Biomimetics, vol. 10, no. 5, p. 055003, 2015.
[13] C. Laschi and M. Cianchetti, ”Soft Robotics: New
Perspectives for Robot Bodyware and Control”,Frontiers
in Bioengineering and Biotechnology, vol. 2, 2014.
[14] P. Polygerinos, S. Lyne, Z. Wang, L. F. Nicolini, B.
Mosadegh, G. M. Whitesides, and C. J. Walsh, “Towards a
soft pneumatic glove for hand rehabilitation,” 2013
IEEE/RSJ International Conference on Intelligent Robots
and Systems, 2013.
[15] B. Radder, G. B. Prange-Lasonder, A. I. R. Kottink, L.
Gaasbeek, J. Holmberg, T. Meyer, J. H. Buurke, and J. S.
Rietman, “Preliminary Findings of Feasibility of a
Wearable Soft-robotic Glove Supporting Impaired Hand
Function in Daily Life – A Soft-robotic Glove Supporting
ADL of Elderly People,” Proceedings of the International
Conference on Information and Communication
Technologies for Ageing
Well and e-Health, 2016.
[16] E. Brown, N. Rodenberg, J. Amend, A. Mozeika, E. Steltz,
M. Zakin, H. Lipson and H. Jaeger, ”Universal robotic
gripper based on the jamming of granular material”,
Proceedings of the National Academy of Sciences, vol. 107,
no. 44, pp. 18809-18814, 2010.

IRJET- Investigation into use of Soft Robotic End Effector for Assistive Gripping

Recommended

Recommended

More Related Content

What's hot

What's hot (18)

Similar to IRJET- Investigation into use of Soft Robotic End Effector for Assistive Gripping

Similar to IRJET- Investigation into use of Soft Robotic End Effector for Assistive Gripping (20)

More from IRJET Journal

More from IRJET Journal (20)

Recently uploaded

Recently uploaded (20)

IRJET- Investigation into use of Soft Robotic End Effector for Assistive Gripping